echelon scan based on the number of regions
e.scan(x, locs, peaks, c_separates, parents, K)
x
.
locs
peaks
c_separates
parents
K